PERMEABILITY BY THE METHOD OF TISSUE TENSION 565 
slightly hypertonic to the cells ; but a very slight shrinkage of the cells, 
such as caused the decrease in curvature of the tissue, sufficed to raise 
their osmotic pressure to equality with that of the solution, and 
recovery proceeded as rapidly as the rate of penetration of the plas- 
molyzing substance would permit. It will be seen that while the 
solutions used were not strictly isotonic with the cells, the latter 
adjusted themselves to the solution, becoming isotonic with it within 
at most a very few minutes. 
As soon as a condition of isotony was established between cells 
and solution, the concentration of the latter was increased by the 
addition of a measured small amount of a molecular solution of the 
salt, and the solution quickly stirred in order to secure a uniform dis- 
tribution of the increase in concentration. In order to avoid very 
great differences in the amount of shrinkage, such as might cause a 
mechanical alteration of the cell walls, or injury to the protoplasm, it 
was necessary to vary this increase in concentration according to the 
salt used. Thus, for the sea water-calcium chloride mixture, described 
by the writer in a previous paper (2), and for the salts of univalent 
kations 0.2 cc. of i ikf solution was added to 20 cc. of the isotonic solu- 
tion which had a concentration of 0.20 to 0.235 M. In the case of 
bivalent kations the concentrations were 0.15 to 0.17 M and to 20 cc. 
of the solution there was added o.i cc. of a i ikf solution to produce 
the desired increase of concentration. In the case of trivalent kations 
the concentrations were .045 to .065 M and 0.05 cc. of a i ikf solution 
was added when it was desired to increase the concentration. The 
maximum error in making these changes was 0.0001 M. 
This increase in concentration resulted in a decrease in the curvature 
of the strip of tissue which soon ceased and was followed by a slower 
movement in the opposite direction. The time elapsing between the 
increase of concentration and the moment when the strip regained 
its original curvature {i. e., returned to its initial position on the scale 
of the ocular micrometer) was recorded as the "time of recoverv." 
Immediately upon the recovery of a given strip, the concentration of 
the solution bathing it was again raised by the same amount as before, 
and the time of recovery again noted. By repetition of this process 
we secure a series of recoveries of one strip of tissue, the time required 
for each recovery being a measure of the average rate of penetration 
of salt during that recovery, and the initial and final curvature being 
always the same throughout the whole of the experiment. 
